Scalable Subject Generation: Enhancing Discovery with UCF’s Custom AI

Scalable Subject Generation: Enhancing Discovery with UCF’s Custom AI

The University of Central Florida (UCF) Libraries are leveraging artificial intelligence (AI), specifically the OpenAI API, to transform their metadata workflows. By automating the assignment of Faceted Application of Subject Terminology (FAST) headings and keywords across both digital and traditional collections, these efforts aim to enhance resource discoverability and improve the overall user experience. The project explores multiple term reconciliation approaches, including the use of the OCLC service and the development of a custom FAST vector database. Through systematic testing and evaluation—such as comparisons with Alma’s AI Metadata Assistant—the Libraries are refining their AI-driven metadata strategies. This work supports a scalable approach to subject generation, enhancing discovery through UCF’s custom AI framework and paving the way for more efficient and enriched library services. Presentation delivered at the Library’s AI Interest Group meeting on January 27, 2026.

Deng, Sai (Sai Sophie)

Piascik, Jeanne

2026-01-27

application/vnd.openxmlformats-officedocument.presentationml.presentation

eng

Text; Presentation